Biosensor arrays and chemosensor arrays serve for detecting molecules in an analyte to be examined. Such arrays are increasingly being realized on chips for the purpose of miniaturization. The sensors are often arranged in a large number on a substrate. The substrate may be a semiconductor chip (silicon) for example, particularly for the case where functions of an integrated electronic circuit are intended to be realized.
Such substrates may alternatively be produced from glass, plastic or another material provided that no active electronics are required for operating them on the substrate. The high degree of parallelization enables a simultaneous parallel implementation of different tests, for example tests for the presence of different substances (e.g. molecules) in a predetermined analyte. On account of this property, such sensor arrangements including a corresponding evaluation system obtain diverse applications in medical diagnosis technology, in the pharmacological industry (e.g. for pharmacological screening, “high throughput screening”, HTS), in the chemical industry, in foodstuffs analysis, and in ecological and foodstuffs technology.
The basic principle of many known sensors is based on the fact that firstly so-called catcher molecules are applied, e.g. using microdispensing techniques, and immobilized in a position-specific manner on a chip.
FIG. 1 shows a sensor arrangement 100 known from the prior art, in which a multiplicity of sensor arrays 102 are arranged in matrix-type fashion on a chip 101. The sensor arrays 102 are arranged in N columns and in M rows, that is to say at N×M positions, different catcher molecules being immobilized on each sensor array.
FIG. 2A to FIG. 2F in each case show a diagrammatic cross-sectional view of one of the sensor arrays 102 shown in FIG. 1. In particular, FIG. 2A to FIG. 2C show a first sensor array 200, and FIG. 2D to FIG. 2F show a second sensor array 201, the different illustrations of the first sensor array 200 in FIG. 2A to FIG. 2C corresponding to three different operating states, and the illustrations in FIG. 2D to FIG. 2F analogously corresponding to three different operating states of the second sensor array 201.
Each of the sensor arrays 200, 201 has a sensor electrode 202 integrated in the chip 101. First catcher molecules 203 are immobilized on the sensor electrode 202 of the first sensor array 200, the first catcher molecules 203 being DNA half strands. Second catcher molecules 204, which differ from the first catcher molecules 203, are immobilized on the sensor electrode 202 of the second sensor array 201.
FIG. 2A, FIG. 2D show the first sensor array 200 and the second sensor array 201, respectively, in an operating state in which the sensor arrangement 100 is free of potential binding partners (e.g. DNA half strands).
For the purpose of diagnosis, that is to say for examining an analyte for the presence of specific DNA molecules, an analyte 205 is firstly applied to all the sensor arrays 102 of the sensor arrangement 100 and therefore also to the sensor arrays 200, 201, i.e. the entire sensor arrangement 100 is flooded with the analyte 205 to be examined. This operating state of the first sensor array 200 is shown in FIG. 2B, and of the second sensor array 201 in FIG. 2E. Since the first catcher molecules 203 fit together with (half-stranded) DNA molecules present in the analyte 205, namely with the particles 206 to be detected, in accordance with the key/lock principle, a hybridization is effected, i.e. a binding of the respective DNA molecules 206 to the complementary first catcher molecules 203 of the first sensor array 200 (cf. FIG. 2B). Since the second catcher molecules 204 do not fit together with the particles 206 to be detected on account of their base sequence (cf. FIG. 2E), no hybridization is effected.
In order to obtain the operating states of the first sensor array 200 and of the second sensor array 201 respectively shown in FIG. 2C, FIG. 2F, the analyte 205 is removed from the sensor arrangement 100. Furthermore, a rinsing solution 207 is applied to the sensor arrangement 100. As a result, the particles 206 to be detected that have hybridized with the first catcher molecules 206 remain on the first sensor array 200, whereas only the second catcher molecules 204, but not particles 206 to be detected, remain on the second sensor array 201.
Optical methods are often used for detecting the hybridization that has taken place.
In an optical method, a fluorescent marker (“label”) is bound to the DNA strands present in the analytes. If the entire sensor arrangement 100 is then irradiated with electromagnetic radiation (e.g. light) after a hybridization process that has taken place and after a further rinsing step, it is possible, on the basis of knowledge of the localization of the respective catcher molecules 203, 204, to determine the sensor arrays at which a hybridization has taken place (first sensor array 200) and the sensor arrays at which hybridization has not taken place (second sensor array 201). On the basis of the precise knowledge of the catcher molecules 203, 204 used, it is possible to deduce the presence or absence of specific particles to be detected in the analyte to be examined with a high selectivity. The optical methods have the disadvantage of needing a complicated and cost-intensive optical system for evaluation. This makes it more difficult for example to use such optical systems in physicians' practices, or in portable devices at all.
As an alternative to the optical method, a hybridization event that has taken place can be detected using an electric method.
In this respect, it is necessary to distinguish between methods based on the use of an enzyme label (described for example in [1], [2]) and so-called “label-free” methods, described for example in [3] to [9]. Label-free methods are more attractive since a method step for providing molecules with a label, which method step is often complicated from a biochemical standpoint, is avoided and a label-free method is therefore simpler in terms of sample preparation and hence less expensive.
However, the operation of an electronic biosensor is difficult to realize, so that, particularly in the case of the electronic label-free methods, hitherto examinations have been implemented only on individual sensors or on very small arrays comprising a stringing together of individual sensors.
Label-free methods known from the prior art are described below.
A first approach is disclosed in [3] to [6]. This approach is described below with reference to FIG. 3A to FIG. 7B.
FIG. 3A, FIG. 3B show an interdigital electrode arrangement 300 in which a first electrode structure 302 and a second electrode structure 303 are applied in a substrate 301, said electrode structures clearly meshing in interdigitated fashion. FIG. 3A shows a plan view of the interdigital electrode arrangement 300 and FIG. 3B shows a cross-sectional view along the section line I-I′ shown in FIG. 3A. The interdigital electrode arrangement 300 contains periodic electrode components—arranged one beside the other—of the electrode structures 302, 303.
In order to explain the principle of the functioning of the interdigital electrode arrangement 300, a first partial region 304 of the interdigital electrode arrangement 300 will be described with reference to FIG. 4A, FIG. 4B.
The first partial region 304 is shown in a first operating state as a cross-sectional view in FIG. 4A and in a second operating state as a cross-sectional view in FIG. 4B.
Catcher molecules 400 are in each case immobilized on the electrode structures 302, 303. Gold material is preferably used for the electrode structures 302, 303, so that the immobilization of the catcher molecules 400 is realized using the particularly advantageous gold-sulfur coupling known from biochemistry, for example by a thiol terminal group (SH group) of the catcher molecules 400 being chemically coupled to the gold electrodes 302, 303.
An electrolytic analyte 401 to be examined, which is again intended to be examined for the presence of particles 402 to be detected (for example specific DNA molecules), is situated above the sensor electrodes 302, 303 during active sensor operation. A hybridization, that is to say a binding of DNA strands 402 to the catcher molecules 400, is effected only when the catcher molecules 400 and the DNA strands 402 match one another in accordance with the key/lock principle (cf. FIG. 4B). If this is not the case, then no hybridization is effected. The specificity of the sensor is thus derived from the specificity of the catcher molecules 400.
The electrical parameter that is evaluated in the case of this measurement is the complex electric resistance, i.e. the impedance 403 between the electrodes 302, 303, which is illustrated diagrammatically in FIG. 4A, FIG. 4B. On account of a hybridization that has taken place, the value of the impedance changes since the DNA particles 402 to be detected and the catcher molecules 400 comprise a material having electrical properties that deviate from the material of the electrolyte and, after the hybridization, the electrolyte is clearly displaced from the volume surrounding the electrodes 302, 303.
FIG. 5 shows a second partial region 305 of the interdigital electrode arrangement 300 in a cross-sectional view. The second partial region 305 represents a larger partial region of the interdigital electrode arrangement 300 than the first partial region 304 illustrated in FIG. 4A, FIG. 4B. FIG. 5 diagrammatically shows the profile of the electric field lines 500 between respectively adjacent electrode structures 302, 303.
As is furthermore shown in FIG. 5 the field profiles are periodic within a respective imaginary region through two lines of symmetry 501, so that the consideration of two directly adjacent electrode structures 302, 303 that is shown in FIG. 4A, FIG. 4B is sufficient. Furthermore, FIG. 5 diagrammatically shows a coverage region 502 for each of the electrode structures 302, 303, said coverage region representing the catcher molecules immobilized on the electrode structures 301, 302 and particles to be detected that have possibly hybridized with said catcher molecules. It can clearly be understood from the illustration shown in FIG. 5 that the profile of the field lines 500 is significantly influenced on account of a hybridization event since the physicochemical properties particularly of the coverage region 502 are altered.
It should furthermore be noted that, supplementarily or alternatively, catcher molecules may be provided in regions between electrodes 302, 303. The electrical properties of the electrodes or the electrode region again change in the case of hybridization events between catcher molecules provided in regions between the electrodes and particles to be detected.
FIG. 6 diagrammatically shows a simplified equivalent circuit diagram 600 of the first partial region 304 of the interdigital electrode arrangement 300 shown in FIG. 4A.
The equivalent circuit diagram 600 shows a variable first capacitance 601 CM, the value of which is dependent on the extent of a hybridization effected at the electrode structure 302. A variable first nonreactive resistance 602 RM is connected in parallel with the capacitance. Clearly, the components 601, 602 represent the electrical properties of the surrounding region of the first electrode structure 302. The diagram furthermore shows a variable second capacitance 603 CE and a variable second nonreactive resistance 604 RE connected in parallel therewith, which represents the electrical properties of the analyte 401.
Moreover, the diagram shows a variable third capacitance 605 CM and a variable third nonreactive resistance 606 RM connected in parallel therewith, representing the electrical properties of the surrounding region of the second electrode structure 303. As is furthermore shown in FIG. 6, the parallel circuit comprising components 601, 602, the parallel circuit comprising components 603, 604 and the parallel circuit comprising components 605, 606 are connected in series. The components 601, 602, 605 and 606 are represented in variable fashion in order to illustrate that their values change on account of a sensor event.
In order to determine the value of the impedance, in one measuring method, an AC voltage Vchar is applied to one of the electrodes 302, 303, as shown in the equivalent circuit diagram 700 of the first partial region 304 shown in FIG. 7A. The AC voltage Vchar is provided using an AC voltage source 702. The current Imeas flowing through the arrangement is detected using the ammeter 701. The components 701, 702 are connected in series with one another and are connected between the parallel circuit comprising components 605, 606 and the electrical ground potential 703. The AC current signal Imeas resulting at the electrodes 302, 303 is evaluated together with the applied AC voltage Vchar in order to determine the impedance. As an alternative, in another measuring method a signal, that is to say an electrical voltage, may also be applied in each case to both electrodes 302, 303, the signals then being in antiphase.
The version of a simplified equivalent circuit diagram 710 shown in FIG. 7B differs from the equivalent circuit diagram 700 shown in FIG. 7A in that the elements CM 601, 605 and RM 602, 606 have been combined to form a first effective capacitance 711 and, respectively, to form a first effective nonreactive resistance 712.
The distance between the electrodes 302, 303 and the width of the electrodes is typically in the sub-μm range. In accordance with the interdigital electrode arrangement 300, a multiplicity of electrode components (clearly fingers) of the electrode structures 302 and 303 are arranged parallel to one another. Circular arrangements are used in [3] to [6] for reasons of surface functionalization. The external dimensions or the diameter of such individual sensors is typically in the range of from approximately 100 μm to several hundreds of μm.
With regard to the exciting AC voltage Vchar, it should be taken into account that its root-mean-square value or its peak value ought not to exceed a specific maximum value. The biochemical or electrochemical boundary conditions enabling the operation of such sensors are violated when such a maximum value is exceeded. If the electrode potential (which is referred to the electrical potential of the electrolyte) exceeds an upper threshold value, then specific substances may be oxidized in a surrounding region of an electrode. If the electrical potential (which is referred to the electrical potential of the electrolyte) falls below a lower threshold value, substances are reduced there.
An undesirable oxidation or reduction may have the effect, inter alia, of breaking up the chemical bonds entered into during immobilization and hybridization. Furthermore, electrolysis may commence at the sensor electrodes, so that the electrolysis products bring the chemical milieu required for operation of the sensors out of the required equilibrium or lead to gas formation. The absolute values of the critical potentials depend on the composition and the concentration ratio and the chemical surroundings of the electrodes (for example an immobilization layer, an analyte, etc.).
Typical values for the exciting voltage lie in the range of a few 10 mV to at most around 100 mV. This is an important boundary condition for the operation of such sensors since the resulting measurement signal (current intensity Imeas), with regard to its magnitude, is approximately directly proportional to the applied voltage.
A second principle of a label-free electrical sensor such as is disclosed in [7] to [9] is described below with reference to FIG. 8 to FIG. 10.
In accordance with this second approach, a planar electrode is in each case used for the detection of a species, that is to say for the immobilization of catcher molecules and for hybridization with particles to be detected. Furthermore, an AC voltage signal is applied directly to an electrically conductive analyte. In the case of these methods, the application of the AC voltage and the optionally required additional application of a DC offset are effected using a so-called counterelectrode or reference electrode, which realizes a low-impedance electrical coupling to the electrolyte, which electrical coupling is always defined under changing electrochemical conditions and is constant in terms of its electrical properties. Such a reference electrode is usually produced from a different material (for example silver/silver chloride) than the electrodes that are utilized for immobilizing the catcher molecules and are therefore often produced from gold material. The use of different materials results from the different electrochemical requirements made of the two electrode materials.
FIG. 8A, FIG. 8B show a sensor arrangement 800 in accordance with this second approach. FIG. 8A shows a plan view of the sensor arrangement 800 and FIG. 8B shows a cross-sectional view along a section line II-II′ from FIG. 8A.
As is shown in FIG. 8A, a plurality of sensor arrays 802 and a common reference electrode 803 are arranged on a silicon substrate 801. Provided on the surface of each sensor array 802 is an active region 805, on which catcher molecules are immobilized, for hybridization with complementary particles to be detected. An analyte 804 is filled into the sensor arrangement 800. The sensor arrangement 800 uses a silicon substrate 801, although the electrical properties of the silicon are not utilized, in order to form powerful integrated electronics therein.
FIG. 9 shows an equivalent circuit diagram 900 of a partial region 806 of the sensor arrangement 800. This shows a variable first capacitance 901 CM, which represents the capacitance of the surrounding region of the sensor array 802. Furthermore, a variable first nonreactive resistance 902 RM connected in parallel therewith is shown, representing the nonreactive resistance of the surrounding region of the sensor array 802. A variable second capacitance 903 CE and a variable second nonreactive resistance RE 904 connected in parallel therewith represent the electrical properties of the analyte 804.
Furthermore, FIG. 10 shows a further equivalent circuit diagram 1000 of the partial region 806 of the sensor arrangement 800. The latter exhibits, in addition to the components shown in FIG. 9, an AC voltage source 1002, by which an AC voltage can be applied, and exhibits an ammeter 1001 for detecting a measurement current Imeas. The components 1001, 1002 connected in parallel are connected between the electrical ground potential 1003 and the parallel circuit comprising components 903, 904.
Often only very small sample volumes are available in biochemistry. In this case, the use of the sensor arrangement 800 is disadvantageous since the counterelectrode 803 can be provided in miniaturized form only in a very complicated manner, or not at all. It is often realized by a small chlorinated silver tube.
In the case of the described sensor arrangements known from the prior art, the problem occurs during operation or evaluation of measurement signals that the impedance between the electrodes does not have exclusively capacitive components, but rather is a relatively complex, composite quantity. A fundamental reason for this is that, at the measurement electrode, that is in direct electrical (galvanic) contact with the electrolyte, an electrochemical conversion always takes place which is at equilibrium only precisely when the electrical potential of the electrode with respect to the electrolyte can be set freely. Any displacement of this electrical potential automatically results in a net conversion of material at the electrodes which, metrologically, is manifested as an approximately ohmic conductivity.
The immobilization of catcher molecules in principle influences the material conversion at the electrode surface since the electrode is partially covered thereby, and on account of specific electrical properties of the molecules (for example on account of the fact that DNA molecules are often present as polyanions). This makes it more difficult for the detected sensor signals to be evaluated metrologically. Therefore, it is attempted to configure the measurement in such a way that only the value of the electrode capacitance CE that is dependent on the hybridization in the equivalent circuit diagrams specified is determined. As an alternative, it is possible to measure magnitude and phase of the impedance as a function of the exciting frequency, so that ideally all parameters can be determined from the resulting Bode diagram. However, this procedure is very complicated.
One possibility for obtaining signals that can be evaluated in an improved manner consists in the use of a so-called lock-in amplifier for detecting the sensor signal. This principle is explained below on the basis of the equivalent circuit diagrams 900, 1000 shown in FIG. 9, FIG. 10.
With the aid of a lock-in measuring device, an AC voltage Vchar with a frequency f is applied to the electrolyte 804 via the counterelectrode 803 which ensures a low-impedance connection to the electrolyte 804. It is then possible to measure the imaginary part and the real part of the complex total current Imeas resulting from the elements CM, RM, CE and RE.
Assuming that the magnitude of the complex impedance component of the electrolyte 804, namely 1/(2πfCE), is significantly greater than the magnitude of the purely resistive component RE, the measured current results as:
                              I          meas                =                              V            char                    ×                      1                                          R                E                            +                                                                    R                    M                                    ×                                      1                                          j                      ⁢                                                                                          ⁢                      2                      ⁢                                                                                          ⁢                      π                      ⁢                                                                                          ⁢                      f                      ⁢                                                                                          ⁢                                              C                        M                                                                                                                                  R                    M                                    +                                      1                                          j                      ⁢                                                                                          ⁢                      2                      ⁢                                                                                          ⁢                      π                      ⁢                                                                                          ⁢                      f                      ⁢                                                                                          ⁢                                              C                        M                                                                                                                                                    (        1        )            
The imaginary part of the current amounts to:
                              Im          ⁡                      (                          I              meas                        )                          =                              V            char                    ×                                    2              ⁢                                                          ⁢              π              ⁢                                                          ⁢              f              ⁢                                                          ⁢                              C                M                                                                                      (                                                                                    R                        E                                                                    R                        M                                                              +                    1                                    )                                2                            +                              4                ⁢                                                                  ⁢                                  π                  2                                ⁢                                  f                  2                                ⁢                                  C                  M                  2                                ⁢                                  R                  E                  2                                                                                        (        2        )            
Under the further assumption that the nonreactive resistance of the electrolyte RE is significantly less than the parasitic sensor parallel resistance RM, that is to say if RM>>RE holds true, and assuming that the frequency f is chosen to be sufficiently low, so that4π2f2CM2RE2<<1  (3)is satisfied, then to an approximation the simple relationshipIm(Imeas)=Vchar×2πfCM  (4)can be specified for equation (2). Equation (4) clearly states that the imaginary part of the current that is determined by means of the lock-in method depends linearly on the sensor capacitance CM.
It is only under these conditions that the precise change of CM comprises the information sought.
The need to satisfy equation (3) sufficiently well upwardly limits the choice of measurement frequency. However, the free choice of a frequency that is not all that low is desirable since in accordance with equation (4) the magnitude of the measurement signal to be evaluated rises proportionally with the frequency. In order to obtain a signal that can be evaluated well in accordance with equation (4) even in the case of the low frequencies and the stipulations for the order of magnitude of the voltage Vchar, it is necessary to use either large-area sensors, which lead to large values for the sensor capacitance CM, or highly sensitive amplifiers, which is complicated.
The use of a lock-in amplifier has a considerable disadvantage despite the improved detection sensitivity. Complex circuit arrangements are required for the driving and evaluation, said circuit arrangements requiring a considerable space requirement despite a miniaturization of the sensor arrangement compared with the arrangements known from the prior art, with the result that only a limited number of said sensor arrangements can be arranged in matrix-type fashion on the chip area.
Document [10] discloses a method for detecting an analyte, wherein an integrated system is provided which has a sensor arrangement having a plurality of different sensors, wherein each sensor has a matrix composed of alternately conductive and nonconductive regions. When the analyte is present, the electrical resistance of the sensor arrangement changes. This temporal change in the electrical resistance is measured and compared with stored data, the requisite memory unit and the processing unit being integrated in the substrate of the sensor arrangement. The temporal profile of the measurement curve of the resistance values is dependent on the permeation of the sensor arrangement by the analyte. This specific property of the analyte is utilized for detection.